Hardware
Run:ai open source comes to Lambda Kubernetes Service
NVIDIA open‑sourced Run:ai in October 2026, and Lambda Kubernetes Service (LKS) is adding it as a first‑class option after supporting the tool on its infrastructure since 2022. Run:ai extends Kubernet
NVIDIA open‑sourced Run:ai in October 2026, and Lambda Kubernetes Service (LKS) is adding it as a first‑class option after supporting the tool on its infrastructure since 2022.
Run:ai extends Kubernetes with GPU sharing and quota‑based scheduling, letting workloads request fractional GPUs and allowing teams to borrow idle capacity from one another; this increases utilization and reduces experiment wait times.
The LKS integration aims to give users finer control over GPU allocation in shared clusters while maintaining node health through auto‑remediation.
Related
- Lambda assembles leadership team to power gigawatt-scale AI infrastructure for the superintelligence era
- Running Large-Scale GPU Workloads on Kubernetes with Slurm
- NVIDIA Run:ai Is Now Open Source as Part of NVIDIA DSX AI Factory Platform
Source: Lambda Labs | 2026-10-13